D
Manufacturer
DeepReinforce
Manufacturer profile and assigned models.
Best token generation
1.799,4 tok/s
Ornith-1.0-35B
Best prefill
28.681,2 tok/s
Ornith-1.5-35B-A3B
∅ Token generation
351,6 tok/s
Average of 166 runs
∅ Prefill
5.191,4 tok/s
Average of 166 runs
Best TTFT
229 ms
Ornith-1.0-9B
Best harness rate
0%
no harness run yet
🏆
Ornith-1.0-35B · fastest model at 1.799,4 tok/s token generation
Best prefill of this manufacturer: 28.681,2 tok/s (Ornith-1.5-35B-A3B)
Best prefill of this manufacturer: 28.681,2 tok/s (Ornith-1.5-35B-A3B)
Top models by token generation
Best measured generation throughput per model (tok/s), published performance benchmarks.
Models
All benchmarks →7 assigned
Ornith-1.0-35B35B39 runs
446,8tok/s Ø
Min 25,0Max 1.799,4
Ornith-1.0-35B-FP835B0 runs
No performance data yet
Ornith-1.0-9B9B97 runs
330,0tok/s Ø
Min 29,6Max 1.560,5
Ornith-1.5-35B-A3B35B18 runs
331,3tok/s Ø
Min 125,5Max 626,4
Ornith-1.5-35B-A3B-FP835B0 runs
No performance data yet
Ornith-1.5-397B397B0 runs
No performance data yet
Ornith-1.5-9B9B12 runs
246,6tok/s Ø
Min 70,6Max 472,2
